Project Summary It is well known that structures in the medial temporal lobe (MTL), including the hippocampus, are necessary for the formation, consolidation and retrieval of new episodic memories. Perhaps the most common neuroscientific approach to understanding the brain systems supporting episodic memory is to examine encoding and retrieval of information presented on single trials. While this approach has yielded significant insights, it overlooks the fact that new experiences often are meaningfully related to prior experiences, and as such, are integrated with existing knowledge so that we may retain and update our internal models of the world. This observation raises the important questions of how we (1) store the details of individual events that are part of a greater whole, (2) extract the structural commonalities across multiple related events that occur over gaps in time, and (3) how knowledge structures new learning. To date, little is known about the neural mechanisms underlying these critical aspects of episodic memory. The long-term goal is to understand the mechanisms that support the development, representation and use of prior knowledge to bolster new episodic learning. The objectives of the current proposal are (1) to test a model of how prior knowledge in the form of prediction about the structure of experience can increase memory encoding by recruiting processes supported by ventromedial PFC during new learning and (2) to test hypotheses about feature-based integration that leads memories with similar features and contexts to evolve to have overlapping cortical representations through consolidation mechanisms. The central aim of the project is to establish the role of vmPFC and other cortical regions in communication with the hippocampus in mediating the benefits of new learning with prior knowledge. The rationale for the proposed research is that a better understanding of how the memories become stabilized and integrated over time will lead to better cognitive therapies to remediate memory loss. Under Aim 1, a variety of measures/techniques ? including behavior, fMRI, magnetoencephalography (MEG) and electrocorticography (ECoG) ? will be used to examine how prior knowledge about the expected structure of an unfolding event influences the neural mechanisms underlying the temporal encoding of event details. Aim 1's experiments will build on this foundational research by modulating the predictability of the sequential structure of a current event to understand how prior knowledge modulates behavior and neural processes associated with event memory. Aim 2 will explore the initial learning of structured knowledge about related events, which involves the integration of related information across distinct episodes. The proposed experiments will extend this work to examine how the representation of commonalities across experiences can emerge with time via memory consolidation.